Arista Networks, Inc.

Advanced Micro Devices Corporation

Rajshekhar Biradar

Naveen Kumar Devaraj

Rahul Anand Sharma

Gagandeep Singh Arneja

Padmanabh Ratnakar

Rahul Kumar Singh

Vijay Rangarajan

Raghava Sivaramu

Vipin Jain

Embodiments herein describe creating multiple packet fragments from a large data chunk that, for example, exceeds a maximum transmission unit (MTU) supported by a network. In one embodiment, a network interface controller or card (NIC) receives a direct memory access (DMA) from a connected host to transmit an IP packet or data using remote direct memory access (RDMA) technologies. The NIC can evaluate the data chunk associated with the DMA request and determine whether it exceeds the MTU for the network. Assuming it does, the NIC determines how many fragments to divide the data chunk into, and can fragment any portion of the data at flexible packet/payload offsets. The NIC can then retrieve the data chunk from host memory fragment-by-fragment, rather than reading the data chunk all at once, generating headers for the fragments, and then transmit them as packet fragments.

Fragmented transfer of data over networks

Systems and methods for selectively generating a telemetry report to calculate a flow or bit rate are disclosed. The methods include calculating a hash value of a packet, using it as a key to query a bloom filter, and obtaining a packet count. If the packet count reaches a trigger value, a telemetry report is generated and sent, along with the ingress timestamp, hash value, and the packet count, to the collector for calculating the flow rate (or bit rate). The collector compares the packet count and ingress timestamp of the packet of the first telemetry report with a second telemetry report, both reports being generated at various trigger values, and calculates the flow rate. If a hash collision is detected, the calculations are voided and an update to the hash function is suggested.

System and method for scalable and accurate flow rate measurement

A central controller in a data network can maintain a set of access control list (ACL) rules that represent traffic and data policies of the data network. The controller can autonomously propagate the set of ACL rules to switches in the data network. Each switch that receives the set of ACL rules can selectively install rules from the set based on criteria such as whether or not a given rule in the set is close to the source and device class.

Centralized management and distributed enforcement of policies for network segmentation

Systems and methods are provided herein for transitioning a supplicant from one virtual local area network (VLAN) to another using a change of authorization (COA) message. This may be accomplished by an authentication server notifying a network device that a host should be granted access to the network, wherein the authentication server authenticates the host using MAC based authentication. Based on this notification and the MAC address of the host, the network device assigns the host to a first VLAN. If the authentication server determines that the host needs to change from the first VLAN to a second VLAN the authentication server generates a COA message, associated with the host, wherein the COA message comprises a VLAN identifier related to the second VLAN. The authentication server transmits the COA message to the network device causing the network device to route traffic to and from the host using the second VLAN.

Systems and methods for changing a supplicant from one virtual local area network to another using a change of authorization message

The disclosure describes processing packets in connection with a traceroute session in an overlay network that includes detecting traceroute probes using static and dynamic rules and using the time to live (TTL) value in a received traceroute probe to compute an outer TTL value. The TTL value (inner TTL) of the received probe is updated based on the number of underlay routers (hops) comprising the underlay network that are detected during the traceroute session. The received probe with its updated TTL value is encapsulated in an outer frame that includes the computed outer TTL value. The number of hops is updated each time an underlay router sends an ICMP time exceeded message.

Traceroute for overlays using dynamic filters

A facility to determine if performance issues between two host computers in a data network includes a central controller identifying endpoints of a flow path between the two hosts. The central controller communicates with endpoint network devices to initiate telemetry tagging traffic on the flow path. A collector receives telemetry communicated in a packet from network devices on the flow path. A network operator can view the collected telemetry to assess whether the performance issue is in the network or not.

Distinguishing network vs server issues when diagnosing application performance problems

Systems and methods for INT telemetry are disclosed. The system selects a subset of flows from a plurality of flows to monitor. Parameters of the selected flows are sent by the management controller to an INT source for creating a watchlist. The INT source analyses an incoming packet against the parameters in the watchlist to determine if the packet belongs to a flow selected for monitoring. If the packet matches any one of the parameters, then the INT source embeds the packet with an IP address of a designated tenant collector and the INT instructions. A designated collector is allocated for each flow, set of flows, or tenant. Regardless of the path taken by the packet, the embedded INT packet contains all the information needed for a downstream network element to send telemetry data without the need for configuring the network element for telemetry.

Scalable in-band telemetry as a service (TAAS)

A packet loop runs between two participating endpoint network devices, and in particular runs in the respective data planes of the endpoint devices. A probe packet is provided to the data plane of an initiating device and is forwarded to the other device to initiate the packet loop. The source and destination addresses in the probe packet are set equal to a common address. Based on the common address, entries in the respective forwarding tables of the endpoint devices are established to point to each other so that the probe packet is forwarded back and forth between the two devices thus sustaining the packet loop. A broken loop indicates a forwarding path failure at which time corrective action to be taken.

Growing community of inventors

Bengaluru, India

Rajshekhar Biradar